From: "Ni, Ray" <ray.ni@intel.com>
To: "Tan, Dun" <dun.tan@intel.com>,
"devel@edk2.groups.io" <devel@edk2.groups.io>
Cc: "Dong, Eric" <eric.dong@intel.com>,
"Kumar, Rahul R" <rahul.r.kumar@intel.com>,
Gerd Hoffmann <kraxel@redhat.com>
Subject: Re: [Patch V5 19/22] UefiCpuPkg: Combine branch for non-present and leaf ParentEntry
Date: Fri, 24 Mar 2023 08:09:38 +0000 [thread overview]
Message-ID: <MN6PR11MB82447465220D86A83E687C528C849@MN6PR11MB8244.namprd11.prod.outlook.com> (raw)
In-Reply-To: <20230324060020.940-20-dun.tan@intel.com>
Reviewed-by: Ray Ni <ray.ni@intel.com>
> -----Original Message-----
> From: Tan, Dun <dun.tan@intel.com>
> Sent: Friday, March 24, 2023 2:00 PM
> To: devel@edk2.groups.io
> Cc: Dong, Eric <eric.dong@intel.com>; Ni, Ray <ray.ni@intel.com>; Kumar,
> Rahul R <rahul.r.kumar@intel.com>; Gerd Hoffmann <kraxel@redhat.com>
> Subject: [Patch V5 19/22] UefiCpuPkg: Combine branch for non-present and
> leaf ParentEntry
>
> Combine 'if' condition branch for non-present and leaf Parent
> Entry in PageTableLibMapInLevel. Most steps of these two condition
> are the same. This commit doesn't change any functionality.
>
> Signed-off-by: Dun Tan <dun.tan@intel.com>
> Cc: Eric Dong <eric.dong@intel.com>
> Cc: Ray Ni <ray.ni@intel.com>
> Cc: Rahul Kumar <rahul1.kumar@intel.com>
> Tested-by: Gerd Hoffmann <kraxel@redhat.com>
> Acked-by: Gerd Hoffmann <kraxel@redhat.com>
> ---
> UefiCpuPkg/Library/CpuPageTableLib/CpuPageTableMap.c | 85
> ++++++++++++++++++++++++++++++++------------------------------------------
> -----------
> 1 file changed, 32 insertions(+), 53 deletions(-)
>
> diff --git a/UefiCpuPkg/Library/CpuPageTableLib/CpuPageTableMap.c
> b/UefiCpuPkg/Library/CpuPageTableLib/CpuPageTableMap.c
> index ad1e263084..2430f1b37c 100644
> --- a/UefiCpuPkg/Library/CpuPageTableLib/CpuPageTableMap.c
> +++ b/UefiCpuPkg/Library/CpuPageTableLib/CpuPageTableMap.c
> @@ -351,68 +351,45 @@ PageTableLibMapInLevel (
> // ParentPagingEntry ONLY is deferenced for checking Present and
> MustBeOne bits
> // when Modify is FALSE.
> //
> -
> - if (ParentPagingEntry->Pce.Present == 0) {
> - //
> - // [LinearAddress, LinearAddress + Length] contains non-present range.
> - //
> - Status = IsAttributesAndMaskValidForNonPresentEntry (Attribute, Mask);
> - if (RETURN_ERROR (Status)) {
> - return Status;
> - }
> -
> - //
> - // Check the attribute in ParentPagingEntry is equal to attribute calculated
> by input Attribue and Mask.
> - //
> - PleBAttribute.Uint64 = PageTableLibGetPleBMapAttribute
> (&ParentPagingEntry->PleB, ParentAttribute);
> - if ((IA32_MAP_ATTRIBUTE_ATTRIBUTES (&PleBAttribute) &
> IA32_MAP_ATTRIBUTE_ATTRIBUTES (Mask))
> - == (IA32_MAP_ATTRIBUTE_ATTRIBUTES (Attribute) &
> IA32_MAP_ATTRIBUTE_ATTRIBUTES (Mask)))
> - {
> - return RETURN_SUCCESS;
> - }
> -
> + if ((ParentPagingEntry->Pce.Present == 0) || IsPle (ParentPagingEntry,
> Level + 1)) {
> //
> - // The parent entry is CR3 or PML5E/PML4E/PDPTE/PDE.
> + // When ParentPagingEntry is non-present, parent entry is CR3 or
> PML5E/PML4E/PDPTE/PDE.
> // It does NOT point to an existing page directory.
> + // When ParentPagingEntry is present, parent entry is leaf PDPTE_1G or
> PDE_2M. Split to 2M or 4K pages.
> + // Note: it's impossible the parent entry is a PTE_4K.
> //
> - ASSERT (Buffer == NULL || *BufferSize >= SIZE_4KB);
> - CreateNew = TRUE;
> - *BufferSize -= SIZE_4KB;
> -
> - if (Modify) {
> - ParentPagingEntry->Uintn = (UINTN)Buffer + *BufferSize;
> - ZeroMem ((VOID *)ParentPagingEntry->Uintn, SIZE_4KB);
> - //
> - // Set default attribute bits for PML5E/PML4E/PDPTE/PDE.
> - //
> - PageTableLibSetPnle (&ParentPagingEntry->Pnle, &NopAttribute,
> &AllOneMask);
> - } else {
> + PleBAttribute.Uint64 = PageTableLibGetPleBMapAttribute
> (&ParentPagingEntry->PleB, ParentAttribute);
> + if (ParentPagingEntry->Pce.Present == 0) {
> //
> - // Just make sure Present and MustBeZero (PageSize) bits are accurate.
> + // [LinearAddress, LinearAddress + Length] contains non-present range.
> //
> + Status = IsAttributesAndMaskValidForNonPresentEntry (Attribute,
> Mask);
> + if (RETURN_ERROR (Status)) {
> + return Status;
> + }
> +
> OneOfPagingEntry.Pnle.Uint64 = 0;
> + } else {
> + PageTableLibSetPle (Level, &OneOfPagingEntry, 0, &PleBAttribute,
> &AllOneMask);
> }
> - } else if (IsPle (ParentPagingEntry, Level + 1)) {
> - //
> - // The parent entry is a PDPTE_1G or PDE_2M. Split to 2M or 4K pages.
> - // Note: it's impossible the parent entry is a PTE_4K.
> - //
> +
> //
> - // Use NOP attributes as the attribute of grand-parents because CPU will
> consider
> - // the actual attributes of grand-parents when determing the memory
> type.
> + // Check if the attribute, the physical address calculated by
> ParentPagingEntry is equal to
> + // the attribute, the physical address calculated by input Attribue and
> Mask.
> //
> - PleBAttribute.Uint64 = PageTableLibGetPleBMapAttribute
> (&ParentPagingEntry->PleB, ParentAttribute);
> if ((IA32_MAP_ATTRIBUTE_ATTRIBUTES (&PleBAttribute) &
> IA32_MAP_ATTRIBUTE_ATTRIBUTES (Mask))
> == (IA32_MAP_ATTRIBUTE_ATTRIBUTES (Attribute) &
> IA32_MAP_ATTRIBUTE_ATTRIBUTES (Mask)))
> {
> - //
> - // This function is called when the memory length is less than the region
> length of the parent level.
> - // No need to split the page when the attributes equal.
> - //
> if ((Mask->Bits.PageTableBaseAddressLow == 0) && (Mask-
> >Bits.PageTableBaseAddressHigh == 0)) {
> return RETURN_SUCCESS;
> }
>
> + //
> + // Non-present entry won't reach there since:
> + // 1.When map non-present entry to present, the attribute must be
> different.
> + // 2.When still map non-present entry to non-present,
> PageTableBaseAddressLow and High in Mask must be 0.
> + //
> + ASSERT (ParentPagingEntry->Pce.Present == 1);
> PhysicalAddrInEntry =
> IA32_MAP_ATTRIBUTE_PAGE_TABLE_BASE_ADDRESS (&PleBAttribute) +
> (UINT64)PagingEntryIndex * RegionLength;
> PhysicalAddrInAttr =
> (IA32_MAP_ATTRIBUTE_PAGE_TABLE_BASE_ADDRESS (Attribute) + Offset)
> & (~RegionMask);
> if (PhysicalAddrInEntry == PhysicalAddrInAttr) {
> @@ -423,17 +400,19 @@ PageTableLibMapInLevel (
> ASSERT (Buffer == NULL || *BufferSize >= SIZE_4KB);
> CreateNew = TRUE;
> *BufferSize -= SIZE_4KB;
> - PageTableLibSetPle (Level, &OneOfPagingEntry, 0, &PleBAttribute,
> &AllOneMask);
> +
> if (Modify) {
> - //
> - // Create 512 child-level entries that map to 2M/4K.
> - //
> PagingEntry = (IA32_PAGING_ENTRY *)((UINTN)Buffer + *BufferSize);
> ZeroMem (PagingEntry, SIZE_4KB);
>
> - for (SubOffset = 0, Index = 0; Index < 512; Index++) {
> - PagingEntry[Index].Uint64 = OneOfPagingEntry.Uint64 + SubOffset;
> - SubOffset += RegionLength;
> + if (ParentPagingEntry->Pce.Present) {
> + //
> + // Create 512 child-level entries that map to 2M/4K.
> + //
> + for (SubOffset = 0, Index = 0; Index < 512; Index++) {
> + PagingEntry[Index].Uint64 = OneOfPagingEntry.Uint64 + SubOffset;
> + SubOffset += RegionLength;
> + }
> }
>
> //
> --
> 2.31.1.windows.1
next prev parent reply other threads:[~2023-03-24 8:10 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-24 5:59 [Patch V5 00/22] Fix issues in CpuPageTableLib duntan
2023-03-24 5:59 ` [Patch V5 01/22] UefiCpuPkg/CpuPageTableLib: Remove unneeded 'if' condition duntan
2023-03-24 6:00 ` [Patch V5 02/22] UefiCpuPkg/CpuPageTableLib: Add check for input Length duntan
2023-03-24 6:00 ` [Patch V5 03/22] UefiCpuPkg/CpuPageTableLib:Initialize some LocalVariable at beginning duntan
2023-03-24 6:00 ` [Patch V5 04/22] UefiCpuPkg/CpuPageTableLib: Fix the non-1:1 mapping issue duntan
2023-03-24 6:00 ` [Patch V5 05/22] UefiCpuPkg/CpuPageTableLib:Clear PageSize bit(Bit7) for non-leaf duntan
2023-03-24 6:00 ` [Patch V5 06/22] UefiCpuPkg/CpuPageTableLib: Fix issue when splitting leaf entry duntan
2023-03-24 6:00 ` [Patch V5 07/22] UefiCpuPkg/MpInitLib: Add code to initialize MapMask duntan
2023-03-24 6:00 ` [Patch V5 08/22] UefiCpuPkg/CpuPageTableLib:Add check for Mask and Attr duntan
2023-03-24 8:06 ` Ni, Ray
2023-03-24 6:00 ` [Patch V5 09/22] UefiCpuPkg/CpuPageTableLib: Add manual test to check " duntan
2023-03-24 8:06 ` Ni, Ray
2023-03-24 6:00 ` [Patch V5 10/22] UefiCpuPkg/CpuPageTableLib:Modify RandomBoolean() in RandomTest duntan
2023-03-24 6:00 ` [Patch V5 11/22] UefiCpuPkg/CpuPageTableLib: Add LastMapEntry pointer duntan
2023-03-24 8:07 ` Ni, Ray
2023-03-24 6:00 ` [Patch V5 12/22] UefiCpuPkg/CpuPageTableLib:Modify RandomTest to check Mask/Attr duntan
2023-03-24 8:07 ` Ni, Ray
2023-03-24 6:00 ` [Patch V5 13/22] UefiCpuPkg/CpuPageTableLib: Enable non-1:1 mapping in random test duntan
2023-03-24 6:00 ` [Patch V5 14/22] UefiCpuPkg/CpuPageTableLib: Add OUTPUT IsModified parameter duntan
2023-03-24 8:08 ` Ni, Ray
2023-03-24 6:00 ` [Patch V5 15/22] UefiCpuPkg/CpuPageTableLib: Modify RandomTest to check IsModified duntan
2023-03-24 6:00 ` [Patch V5 16/22] UefiCpuPkg: Fix IA32 build failure in CpuPageTableLib.inf duntan
2023-03-24 6:00 ` [Patch V5 17/22] UefiCpuPkg: Modify UnitTest code since tested API is changed duntan
2023-03-24 6:00 ` [Patch V5 18/22] UefiCpuPkg/CpuPageTableLib: Add check for page table creation duntan
2023-03-24 6:00 ` [Patch V5 19/22] UefiCpuPkg: Combine branch for non-present and leaf ParentEntry duntan
2023-03-24 8:09 ` Ni, Ray [this message]
2023-03-24 6:00 ` [Patch V5 20/22] UefiCpuPkg/CpuPageTableLib: Enable PAE paging duntan
2023-03-24 8:10 ` Ni, Ray
2023-03-24 6:00 ` [Patch V5 21/22] UefiCpuPkg/CpuPageTableLib: Add RandomTest for " duntan
2023-03-24 6:00 ` [Patch V5 22/22] UefiCpuPkg/CpuPageTableLib: Reduce the number of random tests duntan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-list from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=MN6PR11MB82447465220D86A83E687C528C849@MN6PR11MB8244.namprd11.prod.outlook.com \
--to=devel@edk2.groups.io \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox